Stable expression of fluorescently tagged proteins for studies ... - Nature

2 downloads 0 Views 335KB Size Report
tagged proteins for studies of mitosis in mammalian cells. Patricia Wadsworth1, Nasser M Rusan1,2, U Serdar Tulu1,2 & Carey Fagerstrom1. 1Department of ...
PROTOCOL

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

PUBLISHED IN ASSOCIATION WITH COLD SPRING HARBOR LABORATORY PRESS

Stable expression of fluorescently tagged proteins for studies of mitosis in mammalian cells Patricia Wadsworth1, Nasser M Rusan1,2, U Serdar Tulu1,2 & Carey Fagerstrom1 1Department of

Biology and 2Program in Molecular and Cellular Biology, University of Massachusetts, Amherst, Amherst, Massachusetts 01003, USA. Correspondence should be addressed to P.W. ([email protected]).

Cultured mammalian cells have been a popular and useful system for studying mitosis1–3. In particular, the direct observations of fluorescently tagged cytoskeletal and chromosomal proteins in live cells have furthered our understanding of the dynamics of cell division4–6. Although cultured cells are not suitable for manipulation using genetic methods, they are simply grown and maintained in the laboratory and easily transfected with dominant-negative or constitutively active constructs, and RNA interference can be used to selectively reduce or eliminate the expression of specific genes. Cultured cells are somatic rather than embryonic and have an extended G1 phase of the cell cycle. They are therefore ideal for studies of the regulation of the somatic cell cycle (Figs. 1 and 2). This protocol describes methods for generating cell lines that constitutively express green fluorescent protein (GFP)-tagged proteins. Although transient transfection is relatively fast, the expression level of the GFP fusion protein under study may vary greatly from cell to cell. Establishing permanent cell lines has the advantage that all cells express the chimeric protein at the same level and each cell line can be characterized. Permanent cell lines are especially useful for observation of cellular events that could be altered by overexpression of a GFP fusion protein. Finally, a major challenge for those who study mitosis is the need to keep cells alive and actively mitotic during observations. The protocol therefore also incorporates methods for maintaining healthy, dividing cells in culture and during imaging.

MATERIALS REAGENTS cDNA insert encoding gene of interest Cell line(s) to be transfected 15% dimethyl sulfoxide (DMSO) pCAUTION Growth medium (normal and selective), appropriate for cell culture Growth medium, bicarbonate-buffered and supplemented with fetal calf serum, for selected clones (see Step 13) HEPES buffer (20 mM, pH 7.2) Lipofectamine 2000 (Invitrogen) pEGFP (BD Biosciences) pIRES (BD Biosciences) Phosphate-buffered saline (PBS), calcium- and magnesium-free Trypsin VALAP7: prepare by placing Vaseline petroleum jelly, lanolin and paraffin at a 1:1:1 ratio in a glass or ceramic vessel and heating at a low setting on a hot plate until thoroughly blended. At room temperature the mixture is solid and should be gently warmed before each use. VALAP is flammable. pCAUTION EQUIPMENT Cloning rings (6- or 8-mm outer diameter size; Fisher Scientific) Culture dishes with grid (MatTek) Humidified incubator with an atmosphere of 5% CO2 Microscopes (see Steps 20 and 21)

© Cold Spring Harbor Laboratory Press

NATURE METHODS | VOL.2 NO.12 | DECEMBER 2005 | 981

PROTOCOL

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

Generating stable cell lines

PROCEDURE 1| To generate the GFP fusion protein, insert the corresponding cDNA into the multiple cloning site of a pEGFP or other appropriate fluorescent protein–coding vector, joining the GFP sequence at the amino- or carboxy-terminal end of the target protein as appropriate. If desired, subclone the DNA fragment encoding the GFP fusion protein into the pIRES vector. ▲CRITICAL STEP 2| Transfect the cells with the recombinant vector using Lipofectamine 2000 or another commercially available transfection reagent, according to the manufacturer’s instructions. Incubate the cells for ∼48 h in normal growth medium to allow expression of the resistance gene. 3| Incubate the cells in the appropriate antibiotic–containing medium for 2 weeks, changing the medium frequently (every 1 to 2 d) to remove dead or dying cells. ▲CRITICAL STEP 4| Treat the surviving cells with trypsin and plate them in 100-mm dishes at a dilution such that single cells will give rise to well-separated, individual colonies. Plate a dish or flask at higher cell density, and keep it in reserve in case appropriate clones are not generated from the dilute plating. The dilution that is sufficient to give rise to individual colonies depends on many variables and must be determined empirically. 5| Allow the cells to grow until individual colonies of several hundred cells are present (usually 1–2 weeks). For cells expressing a GFP-tagged cytoskeletal protein, colonies (of varying brightness) can be located using an inverted epifluorescence microscope and their location on the dish marked; alternatively, colonies can be selected randomly without assaying fluorescence (cloning blind). 6| Remove the medium from the cultured cells and wash them twice with calcium- and magnesium-free PBS, removing the final PBS wash. ➨ TROUBLESHOOTING 7| To isolate individual colonies, prepare a set of cloning rings by applying a thin layer of vacuum grease or petroleum jelly to one face. If the cloning rings will not be used right away, set them aside in a tissue culture hood. Alternatively, cloning discs can be used to select colonies on the 100-mm dish. If the cells of interest do not form tight colonies (for example, fibroblasts), they can be cloned by limiting dilution in 96-well plates. 8| Use sterile forceps to place cloning rings over the selected colonies, coated side down, and press them onto the surface of the plate.

982 | VOL.2 NO.12 | DECEMBER 2005 | NATURE METHODS

a

b

c

d

e

f

Figure 1 | Confocal imaging of LLC-PK-1 cells in mitosis. (a–f) A projection of a Z series of images through a metaphase cell; these images were obtained using a Perkin-Elmer spinning disc confocal scanhead attached to a Nikon microscope (top). The same images are shown after deconvolution with AutoQuant software (bottom). Cells were fixed and stained with antibodies to tubulin (a,d); chromosomes were stained with propidium iodide (b,e); merged images are shown in c and f. Bar, 10µm.

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

PROTOCOL

Figure 2 | Mitosis in living LLC-PK-1 cells expressing GFP-tubulin. Images from a complete sequence of mitosis at the indicated timepoints (h:min:s) are shown: (0:00:00), prophase; (0:07:57), prometaphase; (0:18:24), metaphase; (0:45:04), anaphase; (0:59:56), telophase; and (1:36:29), cytokinesis. Images were acquired using a Perkin-Elmer spinning disc confocal attached to a Nikon microscope. The complete movie is available as Supplementary Video 1. Bar, 10 µm.

9| Pipet 50–70 µl of trypsin into each cloning ring. After the cells in the ring have rounded up (usually after ~3 min), add growth medium to the ring, aspirate the cells and place them into an individual well of a 24-well plate. Avoid pipetting up and down too many times. After this point, each clone in the 24-well plate should be considered as a potential cell line; therefore, care should be taken when splitting and growing the cells to prevent intercontamination among different clones. 10| When the cells have grown to sufficient density in the 24-well plate, treat them with trypsin and then plate them in two wells of a 6-well plate, one of which contains a coverslip. Use the cells on the coverslip to check the level of GFP fusion protein expression with high-resolution fluorescence microscopy; propagate colonies expressing an appropriate level of the chimeric protein from the remaining well. Discard colonies that do not express GFP-tagged protein, are too dim for imaging or overexpress the GFP-tagged protein. 11| Determine the level of expression of the GFP fusion protein in the selected clones by probing western blots of cell extracts with antibodies to the tagged protein and determining the percentage of the total protein that is tagged. ➨ TROUBLESHOOTING

Characterization of clones expressing GFP fusion proteins

12| Measure the mitotic index and doubling time8 of the cloned cell lines and compare with control cells to determine whether expression of the GFP fusion protein alters these parameters. In the case of GFP–microtubule-associated protein (MAP) expression, examine transfected cells for possible effects on the organization of microtubules, membranes and other cytoskeletal elements. 13| Grow the selected clones in 5% CO2 at 37 °C in a humid atmosphere. Use a medium recommended by the American Type Culture Collection (ATCC), buffered with bicarbonate and supplemented with 10% fetal calf serum and antibiotics. In some cases, medium that has been mixed 1:1 with OptiMEM (Invitrogen) prepared with 5% fetal calf serum improves cellular morphology. 14| Plate the cells 1–3 d before use on clean, sterile glass coverslips and allow them to grow to the desired density. Maximize the number of cells in mitosis at the time of observation by plating the cells so that they are in exponential growth at the time of the experiment. The plating density must be determined empirically for each cell line. ➨ TROUBLESHOOTING ▲CRITICAL STEP

NATURE METHODS | VOL.2 NO.12 | DECEMBER 2005 | 983

PROTOCOL

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

15| For observation of living cells in mitosis, remove the bicarbonate-buffered medium and replace it with medium that lacks bicarbonate but instead maintains the pH with HEPES buffer (20 mM, pH 7.2). ▲CRITICAL STEP ■ PAUSE POINT For long-term storage, freeze cultured cells in the bicarbonate-buffered medium used for growing cells. Trypsinize cells, centrifuge and recover the cell pellet in a small volume of growth medium containing 15% DMSO and 20% fetal calf serum. Store the frozen cells in liquid nitrogen tanks. Mounting coverslips for imaging

16| Remove a coverslip from the culture dish and blot excess medium onto a Kimwipe. 17| Invert the coverslip onto Parafilm spacers on a clean glass slide. Parafilm spacers are used to hold the coverslip above the slide surface and prevent the cells from getting squashed under the weight of the coverslip. To make the spacers, cut two thin strips of Parafilm (~2 × 20 mm). Place them on the slide, parallel to each other, about a coverslip-width apart. Place a drop of medium between the spacers and carefully lower the coverslip so that it rests on the spacers. 18| Use the tip of a Kimwipe to remove excess medium from the top of the coverslip and the surface of the slide. 19| Seal the coverslip to the slide with VALAP to prevent evaporation of medium during observation on the microscope. Apply molten VALAP with a small paintbrush to seal the edges of the coverslip to the slide’s surface.

Imaging the cells

20| Place the cells expressing GFP-tagged proteins on the stage of a fluorescence microscope. For observations, keep the cells warm, ideally at 37 °C, although some cultured cells continue to divide at lower temperatures9. For short-term studies, temperature can be maintained using an air curtain incubator, which blows warm air at the sealed preparation. A thermistor probe inserted directly into the Rose chamber (Box 1) can be used to monitor the temperature of the preparation during heating. This method is relatively inexpensive and easy to assemble, but the microscope stand and lens (especially if it is an oil immersion lens) act as heat sinks, and the focus may fluctuate as the microscope warms. This fluctuation can be reduced by heating the microscope area for several hours before beginning the experiment. ▲CRITICAL STEP 21| Observe the cells on an inverted microscope using a high numerical aperture (NA) objective lens (100× or 60×, 1.4 NA) and fluorescence filters optimized for GFP excitation and emission. The illuminating light should be electronically shuttered to minimize exposure of the cells to intense fluorescence excitation. A sensitive, cooled charge-coupled device (CCD) camera should be used to collect images. Time-lapse imaging of dynamic cellular processes can be performed using appropriate software packages to automate image collection and display the resulting images in a movie format. ▲CRITICAL STEP

BOX 1 MOUNTING COVERSLIPS Although the slide-coverslip preparations, as described in the protocol, are easy to make and do not require any special equipment, the use of Rose chambers24 for observation and microinjection of living mitotic cells is preferable19. Rose chambers are not commercially available; however, several manufacturers make chambers similar in design, and some of these chambers also can be used for perfusion experiments (Bioptechs; for additional examples,

984 | VOL.2 NO.12 | DECEMBER 2005 | NATURE METHODS

see ref. 14). Another inexpensive and easy-touse alternative to specially designed chambers is to culture cells in a culture dish with a glass bottom. These dishes are available with a gridded bottom (for example, MatTek) for following individual cells over long periods and can be used with an upright microscope with a water immersion objective lens. Glass-bottom dishes are inexpensive, easy to use and suitable for microinjection experiments.

PROTOCOL

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

TROUBLESHOOTING TABLE PROBLEM

SOLUTION

Step 6 Once medium is removed from the cells, there is a loss in viability.

Only a small amount of PBS remains on the plate during the steps after Step 6; thus, care must be taken to avoid killing the cells by drying (with practice, about 20 colonies can be selected from a 100mm plate before drying is a problem).

Step 11 For certain fusions, the proteins analyzed are not full length; in other cases, the expression levels are far too low.

If expression of a full-length GFP fusion is deleterious to the cell, it may be difficult to obtain the desired stable cell line— either no stable transfectants are obtained or only mutated versions (for example, truncations) of the GFP-tagged protein are recovered. Before proceeding with any in-depth analysis, confirm within reasonable limits (by western blot and/or PCR) that observed transfectants do indeed express full-length protein. If problems are observed, changing to a weaker or regulated promoter10 may be necessary to obtain the desired stable cell lines.

Step 14 The goal is to examine mitotic cells, but only a few percent of cells in the culture are undergoing mitosis.

For studies in which a large number of mitotic cells need to be followed or a particular mitotic stage is desired, synchronization can be useful. For example, prophase cells constitute a minority of the mitotic cells on a coverslip, but incubation of coverslips of cells for 90 min in nocodazole (10 µM) followed by 90 min of washout results in a transient increase in prophase cells in the population11. This procedure takes advantage of the observation that microtubule disassembly delays the G2 to M transition in vertebrate cells11. Mitotic shake-off is an effective method to obtain relatively large numbers of cells in mitosis, but this method is limited to cells that round up in mitosis, such as HeLa. For cells that remain adherent during mitosis, thymidine can be used to block cells at the entry into S phase; after removal of the thymidine, the cells progress into G2 and M. The thymidine procedure is most effective when it is repeated (a double thymidine block)12,13. Colcemid can be used after the removal of thymidine to obtain cells synchronously entering but blocked in mitosis13. A disadvantage of these methods is that, depending on the length of the cell cycle, it can take several days to obtain a synchronized population of cells.

CRITICAL STEPS Step 1 The bicistronic expression vector pIRES can be used to generate stable cell lines expressing GFP-tagged proteins. The vector uses an internal ribosome entry site (IRES) derived from the encephalomyocarditis virus (ECMV), which allows two open reading frames to be translated from one mRNA; thus, the target gene and the selection marker are expressed from the same promoter. The target gene is immediately downstream of the immediate early promoter of cytomegalovirus (CMV IE promoter) and the selection marker is downstream of the IRES site. With this vector system, selection is more efficient, presumably because of the linked expression of both genes. Step 3 Use the lowest concentration of antibiotic that kills all control parental cells. Determine this concentration by generating a ‘death curve’ for the parental cell line before transfection. Treat parental cells growing in a 24-well plate with various concentrations of antibiotic for 10 d; use the lowest level of antibiotic that kills all the cells in the well. Step 14 Each coverslip should be hand-washed in Alconox detergent, rinsed exhaustively in running hot water and then rinsed in distilled water. Washed coverslips are stored in 95% ethanol and flame sterilized before use. See ref. 7 for a detailed description of this cleaning technique. An alternative, although less preferred, method for cleaning coverslips is to simply rinse them in 95% ethanol and flame-sterilize before use. NATURE METHODS | VOL.2 NO.12 | DECEMBER 2005 | 985

PROTOCOL

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

Step 15 Maintaining the cells in HEPES-buffered medium for shortterm observations (several hours) causes no detectable adverse effects on the cells and avoids the need to regulate CO2 level during imaging.

a

b

c

d

e

f

Step 20 For long-term studies or for experiments that require more precise temperature control, the microscope can be enclosed in a chamber that surrounds the microscope or the stage region of Figure 3 | Imaging GFP-tubulin in siRNA-treated LLC-PK1-1α cells. (a–c) the microscope. Such chambers LLC-PK-1α cells were cotransfected with siRNA targeting the kinesinare commercially available (for related protein Eg5 and siGLO to identify the successfully transfected cells. example, Buck Scientific) or can Depletion of Eg5 results in monopolar spindles, shown using GFP-tubulin be constructed from Plexiglas or to image microtubules (a), phase contrast to image chromosomes (b) and siGLO fluorescence to verify transfection (c). (d–f) LLC-PK-1α cells were plywood. Chambers that control transfected as in upper panels, fixed and stained for tubulin (d) and Eg5 the CO2 level, as well as the (e). Two cells from the same microscope field: upper cell has not been 14 temperature, are also available . transfected and shows normal bipolar spindle with Eg5 enriched at poles; In addition to air curtain lower cell has been transfected, has a monopolar spindle and lacks Eg5. incubators, heated stages are Merged images are shown in f. Bar, 10µm. available for some microscopes. These can be effective, especially if the observations are not made with oil immersion objectives. If oil immersion is to be used, a heating collar that surrounds the objective lens can also be purchased (Bioptechs) or made, to prevent the lens from acting as a heat sink. A disadvantage of all of these methods is the introduction of strain into the lenses by the repeated cycles of heating and cooling. This problem can be circumvented by keeping the microscope lens in a dry incubator at 37 °C (Bioptechs) when not in use. A final alternative is to place the entire microscope in an environmental room of constant temperature. Step 21 Prophase cells may revert to interphase if they are damaged by light. This may occur because light-induced damage to DNA activates the G2 checkpoint, and the cells return to interphase15. In general, progression through mitosis and overall cell health is achieved by limiting the exposure of the cells to light, especially light of shorter wavelengths. In practice, when viewing cells with tungsten light for phase contrast or differential interference contrast (DIC) microscopy, using green light and/or adding a heat cut filter is sufficient to keep cells actively mitotic. For fluorescence observations, expose cells to the minimum amount of fluorescent light possible and use medium that lacks the pH indicator dye, which is autofluorescent. To reduce photobleaching during fluorescence observations, add the oxygen-scavenging reagent Oxyrase (Oxyrase, Inc.) at a dilution of 1:100 in the medium. COMMENTS Experimental design will dictate the cell line chosen, and often more than one cell line may be suitable for a given study. Rat kangaroo kidney cells (PtK) are a frequent choice for studies of mitosis in cultured cells. PtK cells have the advantages that they remain especially flat throughout mitosis and that they have a small number of chromosomes, facilitating observations of chromosome behavior. Another advantage of studying mitosis in PtK cells is that a considerable body of information regarding mitosis in these cells, including structural analysis, is already available16–18. PtK cells can be transfected using standard procedures, are easily microinjected, and are excellent for both light and electron microscopy. Pig kidney cells (LLC-PK-1; ref. 2) are another epithelial cell line that remains flat throughout mitosis. These cells have more chromosomes than PtKs and a slightly thicker morphology. The cells, however, are very hardy and are easy to microinject19 and transfect at high efficiency (Fig. 1 and Supplementary Video 1 online). African green monkey kidney cells (BS-C-1) are epithelial-like and remain flat throughout mitosis, and they have a large, broad spindle. These cells are also characterized by a prominent centrosome, and have 986 | VOL.2 NO.12 | DECEMBER 2005 | NATURE METHODS

PROTOCOL been used for observations of centrosome behavior in live cells and for removal of the centrosome by cell cutting2. BS-C-1 cells can be transfected, microinjected and used for microsurgery.

© 2005 Nature Publishing Group http://www.nature.com/naturemethods

EXAMPLE OF APPLICATION RNA interference–based analysis can be performed in cultured cell lines21 expressing GFP-tagged proteins. For example, human cells expressing GFP-tubulin have been transfected with siRNA molecules to deplete proteins that regulate microtubule behavior22. Depletion of the targeted protein is verified by western blotting and/or immunofluorescence staining. After depletion of the targeted gene, the dynamics and distribution of microtubules can be examined in living cells. We have performed targeted gene depletion in LLC-PK-1α cells, a nonhuman cell line that expresses GFP-tagged tubulin3, and similar work has been reported for PtK cells23. In the case of LLC-PK-1α cells, which are derived from pig, an extensive collection of expressed sequence tags (ESTs) is available that can be used to obtain the complete gene sequence of cDNA from LLC-PK-1α cells (mRNA and cDNA prepared using commercially available kits) by PCR, using a high-fidelity polymerase. Once the sequence is available, siRNAs can be designed using commercially available algorithms. In some cases, we have found that the pig sequence is a perfect match for the siRNA used to deplete the orthologous gene from human cells. An example is Eg5, a Kinesin 5 family member. Depletion of the protein is confirmed by immunofluorescence and the siRNA transfected cells show the characteristic monopolar phenotype when examined by fluorescence microscopy (Fig. 3). In other cases, the pig sequence is similar, but not identical to, the sequence used for siRNA in human cells. In these cases pig-specific siRNA probes are designed. Thus, the powerful approach of gene knockdowns can be used in pig cells that have the advantage of remaining flat throughout mitosis, facilitating observations of cytoskeletal dynamics. Note: Supplementary information is available on the Nature Methods website. SOURCE This protocol is adapted from of Live Cell Imaging: A Laboratory Manual (eds. Goldberg, R.L. & Spector, D.L.) 491–501, 537–553 (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA, 2005; http://www.cshlpress.com/link/livecelp.htm). For details of cell culture and manipulation, including trypsinizing cells, see Cells: A Laboratory Manual (eds. Spector, D.L., Goldman, R.D. & Leinwand, L.A.) (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA, 1998; http://www.cshlpress.com/link/cell_man.htm). 1.

McIntosh, J.R., Grishchuk, E.L. & West, R.R. Chromosomemicrotubule interactions during mitosis. Annual Rev. Cell and Devel. Bio. 18, 193–219 (2002). 2. Gorbsky, G.J., Sammak, P.J. & Borisy G.G. Microtubule dynamics and chromosome motion visualized in living anaphase cells. J. Cell Biol. 106, 1185–1192 (1988). 3. Rusan, N.M., Fagerstrom, C.J., Yvon, A.C. & Wadsworth P. Cell cycle dependent changes in microtubule dynamics in living cells expressing GFP-tubulin. Mol. Biol. Cell. 12, 971–980 (2001) 4. Rieder, C.L. & Khodjakov, A. Mitosis through the microscope: advances in seeing inside live dividing cells. Science 300, 91–96 (2003). 5. Bloom, K. et al. Using green fluorescent protein fusion proteins to quantitate microtubule and spindle dynamics in budding yeast. Methods Cell Biol. 61, 369–383 (1999). 6. Tulu, U.S., Rusan, N. & Wadsworth, P. Peripheral, noncentrosome–associated microtubules contribute to spindle formation in centrosome containing cells. Curr. Biol. 13, 1894–1899 (2003). 7. Lutz, D.A. & Inoue, S. Techniques for observing living gametes and embryos. Methods Cell Biol. 27, 89–110 (1986). 8. Stein, G.S. et al. Synchronization of normal diploid and transformed mammalian cells. In Cell Biology, a Laboratory Handbook (ed. Celis, J.E.) 282–287 (Academic Press, New York, 1994). 9. Rieder, C.L. Effect of hypothermia (20–25 °C) on mitosis in PtK1 cells. Cell Biol. Int. Rep. 5, 563–573 (1981). 10. Zhu, Z., Zheng, T., Lee, C.G., Homer R.J. & Elias, J.A. Tetracycline-controlled transcriptional regulation systems: advances and application in transgenic animal modeling. Semin. Cell Dev. Biol. 13, 121–128 (2002). 11. Rieder, C. L. & Cole, R. Microtubule disassembly delays the G2-M transition in vertebrates. Curr. Biol. 10, 1067–1070 (2000). 12. Rao, P.N. & Johnson, R.T. Mammalian cell fusion: studies on the regulation of DNA synthesis and mitosis. Nature 225, 159–164 (1970).

13. Telzer, B.R. & Rosenbaum, J.L. Cell cycle-dependent, in vitro assembly of microtubules onto the pericentriolar material of HeLa cells. J. Cell Biol. 81, 484–497 (1979). 14. McKenna, N.M. & Wang, Y.-L. Culturing cells on the microscope stage. Methods Cell Biol. 29, 195–205 (1989). 15. Rieder, C. L. & Cole, R.W. Entry into mitosis in vertebrate somatic cells is guarded by a chromosome damage checkpoint that reverses the cell cycle when triggered during early but not late prophase. J. Cell Biol. 142, 1013–1022 (1998). 16. Mastronarde, D.N., McDonald, K.L., Ding, R. & McIntosh, J.R. Interpolar spindle microtubules in PtK cells. J. Cell Biol. 123, 1475–1489 (1993). 17. McDonald, K.L., O’Toole, E., Ding, R. & McIntosh, J.R. Kinetochore microtubules in PtK cells. J. Cell Biol. 118, 369–383 (1992). 18. McEwen, B.F., Heagle, A.B., Cassels, G.O., Buttle, K.F. & Rieder, C.L. Kinetochore fiber maturation in PtK1 cells and its implications for the mechanisms of chromosome congression and anaphase onset. J. Cell Biol. 137, 1567–1580 (1997). 19. Wadsworth, P. Microinjection of mitotic cells. Methods Cell Biol. 61, 219–231 (1999). 20. Hinchcliffe, E. H., Cham, M., Khodjakov, A. & Sluder, G. Requirement of a centrososomal activity for cell cycle progression through G1 into S phase. Science 291, 1547– 1550 (2001). 21. Elbashir, S.M. et al. Duplexes of 21-nucleotide RNAs mediate RNA interference in mammalian cell culture. Nature 411, 494–498 (2001). 22. Ganem, N.J. & Compton, D.A. The KinI kinesin Kif2a is required for bipolar spindle assembly through a functional relationship with MCAK. J. Cell Biol. 166, 473–478 (2004). 23. Walczak, C.E. & Stout, J.R. Exploring protein function during mitosis in PtK cells using siRNA. Mol. Biol. Cell. 15, 262a (2004). 24. Rose, G.G., Pomerat, C.M., Shindler, T.O. & Trunnel, J.B. A cellophane strip technique for culturing tissue in multipurpose culture chambers. J. Biophys. Biochem. Cytol. 4, 761–764 (1958).

NATURE METHODS | VOL.2 NO.12 | DECEMBER 2005 | 987